Evaluating bathymetric LiDAR accuracy with different sources of reference data

Abstract

In order to provide proof of consistent quality, survey systems, acquisition methods, and procedures for analyzing the resulting data are subject to regular validation. One method of validation is punctual comparison with data for comparison that entitles as reference data. However, as the conditions at the time of data acquisition critically impact each survey, dedicated parallel surveys employing different methods. This can provide insight into discrepancies and relative consistencies, especially for research purposes. In the literature, multiple techniques and sensors have been presented for the acquisition of underwater reference data, such as multi-beam echo sounders, single-beam echo sounders, and different types of pole measurements. Furthermore, bathymetric LiDAR can be deployed from UAVs, helicopters, and aircraft, each entailing specific data resolution and accuracy. Therefore, this study presents four different bathymetric LiDAR datasets (one UAV, two helicopter and one airplane-based), where for each dataset a different type of reference acquisition approach is used appropriate for the individual water bodies (river, ponds, coastal waters). The results of this comparison display the overall alignment of bathymetric LiDAR and reference data with the highest accuracies for UAV data and pole reference measurements. There, the mean normal distance between the LiDAR data and the reference is 0 cm ± 2 cm standard deviation. The highest difference was seen for the Baltic sea dataset, where airplane-based data and single-beam echo sounder reference were used. In this dataset, the mean normal distance between the LiDAR data and reference is −5 cm ± 10 cm standard deviation. In conclusion, the analyzed bathymetric LiDAR datasets show strong consistency with their respective reference measurements, with observed variations primarily influenced by environmental conditions and system configurations.